%0 Thesis
%9 Doctoral
%A Cavazos Guerra, C.d.C.
%B Department of Geography
%D 2011
%F discovery:1322565
%I UCL (University College London)
%P 181
%T Modelling the atmospheric controls and climate impact  of mineral dust in the Sahara Desert
%U https://discovery.ucl.ac.uk/id/eprint/1322565/
%X Mineral dust aerosols play an important role in climate and the Earth's energy  budget. The effect of dust on the radiative forcing is uncertain due to the complexity of  particle properties and the complexity to quantify and discriminate preferential dust  sources. This research considers the potential of two Regional Climate Models (RCM’s):  The Weather Research and Forecasting model (WRF-Chem) and the Regional Climate  Model (RegCM3) both with an integrated dust module. Numerical sensitivity experiments  are performed to quantify the ability of both models to simulate sources, the magnitude of  dust emission, the transport in 3-dimensions and the subsequent impact on the radiative  forcing. Particular emphasis is given to preferential source regions within the Sahara and  Sahel in North Africa including the Bodélé Depression in Northern Chad. To account for  the distribution of preferential dust source regions, soil texture characteristics were  modified in dust source regions in RegCM3. As for WRF-Chem GOCART scheme, a new  higher resolution erodible fraction map is tested. Moreover, the sensitivity of the results to  the specification of aerosol optical properties to evaluate the impacts of optical  characteristics on the radiative forcing was considered for the RegCM3. Finally, model  outputs are compared to in-situ data: weather stations (WMO) and AERONET and  satellite estimates: MODIS, MISR, OMI, CALIPSO and SEVIRI. Results show that both  models represent the space/time structure of near-surface meteorology well. The tuning of  preferential dust sources tested in this research provides a more realistic representation of  local dust sources, emissions and resulting AOT. This suggest that in the absence of truly  accurate soil maps at high resolution, further refinements to preferential sources map and  its implementation in dust models can lead to useful improvements in simulation of dust  processes and dust forecast accuracy.